Microsoft’s AI business is booming. A filing shows most of it is just OpenAI

Microsoft’s AI momentum continues, but a new filing highlights concentration risk: it booked $24.1B in revenue from OpenAI over the year to June, per Bloomberg. While the AI business appears to be driving strong results (shares jumped after earnings), reliance on a single partner adds uncertainty around sustainability. Net impact is likely more meaningful for investor sentiment than immediate market-wide repricing.

Analysis

This is less a revenue story than a quality-multiple story. When one partner drives a meaningful share of AI monetization, the market should start haircutting the durability of the AI annuity and applying a higher concentration discount to MSFT’s growth premium. The immediate read-through is not a near-term earnings miss, but a lower ceiling on multiple expansion if investors conclude the AI “platform” is still economically dependent on a single counterparty rather than a broad enterprise adoption curve.

Second-order, the leverage shifts toward the partner and toward rival stacks. If OpenAI can represent a large enough share of the economics, it has negotiating leverage on pricing, compute commitments, and distribution terms, which can pressure Microsoft’s margin mix over the next 6-18 months. At the same time, AWS, GOOGL, and ORCL can pitch diversification and redundancy to enterprise buyers who now have a visible example of concentration risk; that helps them in procurement battles even if it does not show up in top-line data immediately.

The key catalyst path is not days, it is the next 1-2 earnings cycles: if Microsoft does not broaden AI revenue disclosure with stronger Copilot monetization, Azure AI attach, or a wider customer base, the stock may struggle to justify a premium versus mega-cap software peers. The thesis breaks if management can show AI revenue diversification accelerating faster than capex growth, or if OpenAI-related economics prove temporary and non-recurring rather than structural.

Key Decisions for Investors

• Use post-earnings strength to buy 1-3 month MSFT put spreads, targeting a modest downside hedge rather than a home-run short; the risk/reward is best if the market starts discounting AI concentration before the next guidance update.
• Pair trade over 3-6 months: short MSFT / long GOOGL or AMZN on the thesis that enterprise AI buyers will pay up for multi-vendor redundancy, while MSFT faces a concentration haircut. Cover if MSFT proves AI revenue diversification in the next quarter.
• If already long MSFT, sell covered calls into strength and treat the position as a slower-growth quality compounder until AI revenue is shown to be broader than a single partner. This is a volatility monetization trade, not a fundamental bearish call.
• Set a watch item on Azure AI/Copilot monetization and any change in partner economics at the next earnings print; if AI revenue remains concentrated, the multiple risk is months, not days.